Coatue's Lucas Swisher: AI capex demand is real, stablecoins are crypto's second killer use case, and quality is the only ZERP lesson that matters

Summary

Lucas Swisher of Coatue enters the conversation broadly optimistic on markets despite a turbulent start to 2025. The NASDAQ and S&P posted their worst start since the NASDAQ's inception in 1971 through April 18th, but a string of positive data has shifted sentiment: tariffs have not driven inflation as feared, the US government ran a budget surplus in June, and PPI came in below every economist's forecast. The core uncertainty remains whether tariff effects on inflation are simply delayed, a question Swisher acknowledges economists cannot confidently answer.

Quality Over Multiples: The ZERP Lesson

The primary takeaway Swisher draws from the 2020-2021 ZERP era is deceptively simple: quality is the only variable that reliably compounds. He uses Ramp as an example of an asset that consistently looked expensive, trading at 50 to 200 times gross profit, yet rewarded patient investors because the underlying business quality was exceptional. The failure mode of that era was growth investors chasing venture rounds with growth dollars and backing too many undifferentiated companies in sectors they did not understand. That pattern, he notes, is beginning to re-emerge in pockets of today's market, including companies raising at $10 billion-plus valuations with no revenue and no product.

AI Capex Demand Is Real and Binding

Swisher is unambiguous on infrastructure: compute constraints are real and are actively limiting product launches. Anthropic cannot release new products because it cannot serve the demand that already exists. The publicly documented friction between Microsoft and OpenAI over compute access reinforces the point. Coatue was a significant investor in CoreWeave ahead of its IPO and remains bullish on the infrastructure buildout. Swisher frames the AI doomer versus AI optimist divide as the defining debate inside hedge fund buildings on the East Coast, contrasting sharply with near-universal optimism in Silicon Valley.

Stablecoins as Crypto's Second Killer Use Case

Coatue views stablecoins as the second major global use case for crypto, after Bitcoin as a store of value or digital gold. The firm is actively working out how to position. Swisher flags Stripe's acquisition of Bridge as a standout strategic move and sees meaningful international opportunity for companies like Ramp in cross-border money movement. No major new pure-play stablecoin startups have captured Coatue's attention yet; the current read is that the value accrues to larger, more established financial infrastructure players.

Enterprise AI Adoption Is Still Early, Coding Leads

Enterprise adoption follows the same historical pattern seen in mobile and messaging: consumers move first, enterprises follow slowly. Coding is the clearest current enterprise value driver, with $1.3 billion in coding ARR added in the last 12 months by startups alone, outside of Anthropic. Legal, via Harvey; search, via Glean; and medicine, via Open Evidence, represent emerging pockets. Swisher broadly agrees with the Tyler Cowen framing that AGI's economic impact will be gradual, constrained by the stickiness of enterprise processes and physical-world dependencies.

Robotics: Too Early for Growth Mandates, Elon Is the Exception

Coatue's growth investing mandate requires clear business models and established category leaders, criteria robotics cannot yet meet. Coatue has a venture-stage position in Skilled, a Carnegie Mellon spinout, but is deliberately waiting for the market to mature before deploying growth capital. The one exception Swisher carves out is Tesla and Elon Musk, who combines xAI's research capacity, Tesla's real-world data, and SpaceX's engineering culture in a way that makes a counter-thesis difficult to construct.

Autonomous Vehicles: Waymo Economics Work, Uber's Future Is Uncertain

Waymo's unit economics are already functional at current scale, and Swisher expects them to improve materially as the vehicle fleet moves beyond Jaguar-based hardware. The most consequential unresolved question for public market investors is what happens to Uber as Tesla builds its own network and Waymo expands its own. Both a bull and bear case exist for Uber, and the market has not resolved which applies. Second-order effects, including what consumers do with two hours of daily commute time recaptured, are seen as meaningful but still speculative.

AI Talent War Will Intensify

Swisher characterizes the surge in AI researcher compensation as a straightforward supply-demand imbalance. The global pool of researchers that major labs genuinely want numbers in the hundreds, while dozens of well-capitalized companies compete for them. He draws a direct parallel to the rise of pod-based hedge funds, which similarly compressed career timelines and inflated compensation in a talent-scarce environment. His expectation is that the intensity of the competition increases rather than moderates.